首页 | 本学科首页   官方微博 | 高级检索  
     

基于LRU的大流检测算法
引用本文:王洪波,裴育杰,林宇,程时端,金跃辉.基于LRU的大流检测算法[J].电子与信息学报,2007,29(10):2487-2492.
作者姓名:王洪波  裴育杰  林宇  程时端  金跃辉
作者单位:北京邮电大学网络与交换技术国家重点实验室,北京,100876
基金项目:国家自然科学基金 , 国家重点基础研究发展计划(973计划) , CNGI
摘    要:高速网络中,检测大流是进行准确流量测量的一种重要可扩展解决方案。该文提出了一种新的基于LRU大流检测算法。它通过引入小流早期丢弃和大流预保护机制以提高测量准确性。算法分析表明:新算法具有10Gbps线速处理能力。该文基于实际互联网数据进行了实验比较,结果显示:与已有算法相比,新算法具有更高的测量准确性和实用性。

关 键 词:流量测量  大流  重尾分布  最近最久未使用
文章编号:1009-5896(2007)10-2487-06
收稿时间:2006-5-8
修稿时间:2006-05-08

A LRU Based Algorithm for Identifying and Measuring Large Flows
Wang Hong-bo,Pei Yu-jie,Lin Yu,Cheng Shi-duan,Jin Yue-hui.A LRU Based Algorithm for Identifying and Measuring Large Flows[J].Journal of Electronics & Information Technology,2007,29(10):2487-2492.
Authors:Wang Hong-bo  Pei Yu-jie  Lin Yu  Cheng Shi-duan  Jin Yue-hui
Affiliation:State Key Laboratory of Networking and Switching, Beijing University of Posts & Telecommunications, Beijing 100876, China
Abstract:Identifying and measuring large flows is an important scalable solution for traffic measuring accurately on high-speed networks.A new algorithm based on LRU replacement scheme is proposed,which uses the policies of"early dropping small flows"and"preparatively protecting large flows"to enhance the accuracy of traffic measurement.An analysis demonstrates that the new algorithm can support the 10Gbps line-speed processing. Experiments are also conducted based on real network traces.Results show that the proposed method is more accurate and practicable than existing algorithms.
Keywords:Traffic measurement  Large flows  Heavy tailed distribution  Least Recently Used (LRU)
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《电子与信息学报》浏览原始摘要信息
点击此处可从《电子与信息学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号